Stats Explained
08.01.05
Let’s start with the most basic stats to measure the ability of a team’s offense and defense.
Offensive efficiency
Points scored per 100 offensive possessions.
Defensive efficiency
Points allowed per 100 defensive possessions.
In order to compute efficiency, we need to know how to compute possessions.
Possessions
We can estimate possessions very well from box score stats by using this formula.
FGA-OR+TO+0.475xFTA
For each team, possessions are counted for the team and their opponents, and then averaged.
Efficiency gives us broad view of how well the offense or defense functions, but we can break efficiency into what Dean Oliver dubbed the Four Factors. Shooting, rebounding, turnovers, and free throws provide the basic components of efficiency.
Effective field goal percentage (eFG%)
(FGM + 0.5*3PM) / FGA
Shooting is measured by effective field goal percentage, which differs from conventional field goal percentage by taking into account the extra value of a made 3-pointer.
Offensive rebounding percentage
OR / (OR + DR)
Defensive rebounding percentage can also be computed, using defensive rebounds in the numerator.
Turnover percentage
TO / Possessions
Free throw rate
This can either be FTM/FGA or FTA/FGA. Typically, for team offense FTM/FGA is used, while on defense FTA/FGA is used. This is FTA/FGA for both offense and defense.
There are other team stats that are less important than the Four Factors, with the common approach of converting the standard per-game stats to per-opportunity.
Assist Rate
A / FGM
Block Rate
Blocked shots / Opp. 2PA
Steal Rate
Steals / Defensive possessions
All of the above stats can apply to individuals in some form, also. There are two other stats that are applied to individuals that aren’t applied to teams. These stats were developed by Dean Oliver, and the formulas are far too complicated to list here. His book, Basketball on Paper, is worth buying if you are interested in how the calculations are performed.
Offensive Rating
This is the personal version of team offensive efficiency.
Usage (% of possessions used)
This describes a player’s role in the offense, by explaining how many of his team’s possessions a player is personally responsible for ending while he is on the floor.
A simpler version of personal efficiency is this one
Points per weighted shot (PPWS)
Points scored / (FGA + 0.475*FTA)
Basketball Shrink
The Basketball Shrink finds teams most similar to the team you selected, based on the stats on the page you are looking at. On the summary page, teams are compared by adjusted tempo, adjusted offensive efficiency, and adjusted defensive efficiency. On the other stats pages teams are compared by all the stats listed on the page. The algorithm takes the sum of the percentage difference of each stat and ranks the teams accordingly.
I call this the Basketball Shrink because it’s designed to give the user an idea of the personality of a team by producing a list of teams that play a similar style. For more information, here is my original post on the Shrink from the 2005 season: Basketball Shrink Debuts.
What Goes On Here
06.03.05
Back in the day - which was like three years ago - I read a column by Rob Neyer. Rob is a baseball columnist for ESPN.com, and despite that I was not much of a baseball fan, I was addicted to his work. In March of 2002, he wrote what I believe to be his only column about college basketball. The subject was the Kansas Jayhawks’ poor first round showing in the NCAA tourney, an 11-point victory over 16th-seeded Holy Cross that felt closer than the final margin indicated.
I am a big fan of Rob’s because his sensible approach to baseball analysis is refreshing. The way he supports assertions not by talking to a couple of “baseball people,” but by using cold hard facts made me a religious reader of his before ESPN.com charged folks to see his work. Rob’s piece on KU was in this same style, pointing out that based on history, the close game more than likely signaled an early exit for the top-seeded Jayhawks, rather than a wake-up call. That column made me wonder why there wasn’t someone writing stuff like that regularly for college hoops.
You see, college basketball provides a forum for any opinion to be supported. You can claim that because Player X leads his team in scoring, that his teammates must be worse off without his presence. You can think expanding the lane would make the game better. And you can believe that Team X has a bad defense because they allow a lot of points. You can believe all of these things, but around here, I like to test such opinions with the mountains of data that this game with its 300+ teams provide us.
So that’s what propelled me to start a blog in the fall of 2002. My site, kenpom.com, didn’t get much traffic back then, and I made sure to make the link to the blog obscure. Which was why I got about 50 hits a week at the peak of 2003 season. That was good because the content was truly pathetic. The problem was I put the cart before the horse. While I wanted to approach college hoops from a fresh angle, I didn’t have the resources or mindset to do so. Tragically, much like Neyer’s Kansas column, there is no trace of the premier season of my blog to be found on the ‘net.
However, you can look at the archives on this site and see that it was a still a struggle for me for most of the 2004 season. Around tourney time in 2004, I figured out a way to compute the efficiency statistics that are featured on this site. That allowed me to do some objective analysis with revealing data that was previously unavailable to me. For instance, it allowed me to see that Saint Joe’s was legit. Reading Dean Oliver’s Basketball on Paper the following summer sparked a whole lot of new thoughts, and this space has progressed from there.
From the beginning the mission of the blog has been simple - provide something useful. That first season I didn’t accomplish much, but the blog has evolved to provide the occasional nugget that hasn’t been presented anywhere else. Statistical data on rules experiments has been one favorite of mine (with more to come in 2006). More than anything, I want to provide a unique examination of the game. If it means turning conventional wisdom on its ear, all the better.